Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +532 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,532 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 70,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02857142857142857,
|
| 13 |
+
"grad_norm": 0.03305287514367692,
|
| 14 |
+
"learning_rate": 7.142857142857143e-06,
|
| 15 |
+
"loss": 0.1788,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05714285714285714,
|
| 20 |
+
"grad_norm": 0.031936320817185844,
|
| 21 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 22 |
+
"loss": 0.1845,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08571428571428572,
|
| 27 |
+
"grad_norm": 0.04074863585272115,
|
| 28 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 29 |
+
"loss": 0.1837,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11428571428571428,
|
| 34 |
+
"grad_norm": 0.035864053585471116,
|
| 35 |
+
"learning_rate": 2.857142857142857e-05,
|
| 36 |
+
"loss": 0.1851,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.14285714285714285,
|
| 41 |
+
"grad_norm": 0.03092198126963985,
|
| 42 |
+
"learning_rate": 3.571428571428572e-05,
|
| 43 |
+
"loss": 0.1784,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17142857142857143,
|
| 48 |
+
"grad_norm": 0.038462538224975085,
|
| 49 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 50 |
+
"loss": 0.1771,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2,
|
| 55 |
+
"grad_norm": 0.03888233843876885,
|
| 56 |
+
"learning_rate": 5e-05,
|
| 57 |
+
"loss": 0.175,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.22857142857142856,
|
| 62 |
+
"grad_norm": 0.043857940264053076,
|
| 63 |
+
"learning_rate": 4.996892303047306e-05,
|
| 64 |
+
"loss": 0.1691,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2571428571428571,
|
| 69 |
+
"grad_norm": 0.042294142827267324,
|
| 70 |
+
"learning_rate": 4.987576938413504e-05,
|
| 71 |
+
"loss": 0.1669,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2857142857142857,
|
| 76 |
+
"grad_norm": 0.030998439097041534,
|
| 77 |
+
"learning_rate": 4.972077065562821e-05,
|
| 78 |
+
"loss": 0.1602,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3142857142857143,
|
| 83 |
+
"grad_norm": 0.03425972383844138,
|
| 84 |
+
"learning_rate": 4.9504312196213596e-05,
|
| 85 |
+
"loss": 0.1511,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.34285714285714286,
|
| 90 |
+
"grad_norm": 0.03949318202067683,
|
| 91 |
+
"learning_rate": 4.922693215572695e-05,
|
| 92 |
+
"loss": 0.1398,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37142857142857144,
|
| 97 |
+
"grad_norm": 0.039318951731654166,
|
| 98 |
+
"learning_rate": 4.888932014465352e-05,
|
| 99 |
+
"loss": 0.1395,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4,
|
| 104 |
+
"grad_norm": 0.036179419950033265,
|
| 105 |
+
"learning_rate": 4.849231551964771e-05,
|
| 106 |
+
"loss": 0.1341,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.42857142857142855,
|
| 111 |
+
"grad_norm": 0.03497982080054259,
|
| 112 |
+
"learning_rate": 4.803690529676019e-05,
|
| 113 |
+
"loss": 0.1319,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.45714285714285713,
|
| 118 |
+
"grad_norm": 0.03504652911611438,
|
| 119 |
+
"learning_rate": 4.752422169756048e-05,
|
| 120 |
+
"loss": 0.1184,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4857142857142857,
|
| 125 |
+
"grad_norm": 0.03366289822931868,
|
| 126 |
+
"learning_rate": 4.6955539334255716e-05,
|
| 127 |
+
"loss": 0.1248,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5142857142857142,
|
| 132 |
+
"grad_norm": 0.030373551018539273,
|
| 133 |
+
"learning_rate": 4.6332272040803895e-05,
|
| 134 |
+
"loss": 0.1089,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5428571428571428,
|
| 139 |
+
"grad_norm": 0.030313054901733562,
|
| 140 |
+
"learning_rate": 4.5655969357899874e-05,
|
| 141 |
+
"loss": 0.1076,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5714285714285714,
|
| 146 |
+
"grad_norm": 0.030895918951532413,
|
| 147 |
+
"learning_rate": 4.4928312680573064e-05,
|
| 148 |
+
"loss": 0.1092,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.6,
|
| 153 |
+
"grad_norm": 0.029353647238654513,
|
| 154 |
+
"learning_rate": 4.415111107797445e-05,
|
| 155 |
+
"loss": 0.0981,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.6285714285714286,
|
| 160 |
+
"grad_norm": 0.02984852008137626,
|
| 161 |
+
"learning_rate": 4.332629679574566e-05,
|
| 162 |
+
"loss": 0.0957,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6571428571428571,
|
| 167 |
+
"grad_norm": 0.029341330325013314,
|
| 168 |
+
"learning_rate": 4.245592045215182e-05,
|
| 169 |
+
"loss": 0.0854,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.6857142857142857,
|
| 174 |
+
"grad_norm": 0.028162425011642102,
|
| 175 |
+
"learning_rate": 4.154214593992149e-05,
|
| 176 |
+
"loss": 0.0811,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.7142857142857143,
|
| 181 |
+
"grad_norm": 0.027680225358496686,
|
| 182 |
+
"learning_rate": 4.058724504646834e-05,
|
| 183 |
+
"loss": 0.0814,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.7428571428571429,
|
| 188 |
+
"grad_norm": 0.02576777114993237,
|
| 189 |
+
"learning_rate": 3.959359180586975e-05,
|
| 190 |
+
"loss": 0.0762,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7714285714285715,
|
| 195 |
+
"grad_norm": 0.02870619373322991,
|
| 196 |
+
"learning_rate": 3.856365659664399e-05,
|
| 197 |
+
"loss": 0.0687,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8,
|
| 202 |
+
"grad_norm": 0.024863648794319133,
|
| 203 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 204 |
+
"loss": 0.0661,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.8285714285714286,
|
| 209 |
+
"grad_norm": 0.026291877927815625,
|
| 210 |
+
"learning_rate": 3.6405266433829075e-05,
|
| 211 |
+
"loss": 0.0706,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.8571428571428571,
|
| 216 |
+
"grad_norm": 0.025586933701267463,
|
| 217 |
+
"learning_rate": 3.5282177578265296e-05,
|
| 218 |
+
"loss": 0.0645,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.8857142857142857,
|
| 223 |
+
"grad_norm": 0.025589408223466822,
|
| 224 |
+
"learning_rate": 3.413352560915988e-05,
|
| 225 |
+
"loss": 0.06,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.9142857142857143,
|
| 230 |
+
"grad_norm": 0.031037343916879803,
|
| 231 |
+
"learning_rate": 3.2962166256292113e-05,
|
| 232 |
+
"loss": 0.0599,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.9428571428571428,
|
| 237 |
+
"grad_norm": 0.023310202432820483,
|
| 238 |
+
"learning_rate": 3.177101170357513e-05,
|
| 239 |
+
"loss": 0.0569,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.9714285714285714,
|
| 244 |
+
"grad_norm": 0.024283943525703268,
|
| 245 |
+
"learning_rate": 3.056302334890786e-05,
|
| 246 |
+
"loss": 0.0602,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.0,
|
| 251 |
+
"grad_norm": 0.027329875987439457,
|
| 252 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 253 |
+
"loss": 0.0589,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.0285714285714285,
|
| 258 |
+
"grad_norm": 0.02501378282790971,
|
| 259 |
+
"learning_rate": 2.8108592616187133e-05,
|
| 260 |
+
"loss": 0.0543,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.0571428571428572,
|
| 265 |
+
"grad_norm": 0.027119458807348496,
|
| 266 |
+
"learning_rate": 2.686825233966061e-05,
|
| 267 |
+
"loss": 0.0517,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.0857142857142856,
|
| 272 |
+
"grad_norm": 0.02410634194975642,
|
| 273 |
+
"learning_rate": 2.5623267293451826e-05,
|
| 274 |
+
"loss": 0.0495,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.1142857142857143,
|
| 279 |
+
"grad_norm": 0.02638687718523104,
|
| 280 |
+
"learning_rate": 2.4376732706548183e-05,
|
| 281 |
+
"loss": 0.047,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.1428571428571428,
|
| 286 |
+
"grad_norm": 0.024038928955062953,
|
| 287 |
+
"learning_rate": 2.3131747660339394e-05,
|
| 288 |
+
"loss": 0.0453,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.1714285714285715,
|
| 293 |
+
"grad_norm": 0.02760307457120193,
|
| 294 |
+
"learning_rate": 2.189140738381288e-05,
|
| 295 |
+
"loss": 0.0459,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.2,
|
| 300 |
+
"grad_norm": 0.02622815024441306,
|
| 301 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 302 |
+
"loss": 0.0439,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.2285714285714286,
|
| 307 |
+
"grad_norm": 0.035002615228578664,
|
| 308 |
+
"learning_rate": 1.9436976651092144e-05,
|
| 309 |
+
"loss": 0.0491,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.2571428571428571,
|
| 314 |
+
"grad_norm": 0.024513705968269937,
|
| 315 |
+
"learning_rate": 1.8228988296424877e-05,
|
| 316 |
+
"loss": 0.0384,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.2857142857142856,
|
| 321 |
+
"grad_norm": 0.030311524271095066,
|
| 322 |
+
"learning_rate": 1.7037833743707892e-05,
|
| 323 |
+
"loss": 0.04,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.3142857142857143,
|
| 328 |
+
"grad_norm": 0.024027406130505052,
|
| 329 |
+
"learning_rate": 1.5866474390840125e-05,
|
| 330 |
+
"loss": 0.0423,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.342857142857143,
|
| 335 |
+
"grad_norm": 0.021786340454075813,
|
| 336 |
+
"learning_rate": 1.4717822421734718e-05,
|
| 337 |
+
"loss": 0.0361,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.3714285714285714,
|
| 342 |
+
"grad_norm": 0.02644652055974124,
|
| 343 |
+
"learning_rate": 1.3594733566170926e-05,
|
| 344 |
+
"loss": 0.0412,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.4,
|
| 349 |
+
"grad_norm": 0.020152919701181566,
|
| 350 |
+
"learning_rate": 1.2500000000000006e-05,
|
| 351 |
+
"loss": 0.0394,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.4285714285714286,
|
| 356 |
+
"grad_norm": 0.019656157246220014,
|
| 357 |
+
"learning_rate": 1.1436343403356017e-05,
|
| 358 |
+
"loss": 0.0402,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.457142857142857,
|
| 363 |
+
"grad_norm": 0.030741595634745503,
|
| 364 |
+
"learning_rate": 1.0406408194130259e-05,
|
| 365 |
+
"loss": 0.0399,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.4857142857142858,
|
| 370 |
+
"grad_norm": 0.01956578251894237,
|
| 371 |
+
"learning_rate": 9.412754953531663e-06,
|
| 372 |
+
"loss": 0.0371,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.5142857142857142,
|
| 377 |
+
"grad_norm": 0.020137900808619075,
|
| 378 |
+
"learning_rate": 8.45785406007852e-06,
|
| 379 |
+
"loss": 0.0408,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.5428571428571427,
|
| 384 |
+
"grad_norm": 0.026267476897493938,
|
| 385 |
+
"learning_rate": 7.5440795478481815e-06,
|
| 386 |
+
"loss": 0.0361,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.5714285714285714,
|
| 391 |
+
"grad_norm": 0.019075496865514192,
|
| 392 |
+
"learning_rate": 6.673703204254347e-06,
|
| 393 |
+
"loss": 0.0369,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.6,
|
| 398 |
+
"grad_norm": 0.022200908222749335,
|
| 399 |
+
"learning_rate": 5.848888922025553e-06,
|
| 400 |
+
"loss": 0.0341,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.6285714285714286,
|
| 405 |
+
"grad_norm": 0.024122181450616925,
|
| 406 |
+
"learning_rate": 5.071687319426946e-06,
|
| 407 |
+
"loss": 0.0361,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.657142857142857,
|
| 412 |
+
"grad_norm": 0.0186563328727714,
|
| 413 |
+
"learning_rate": 4.344030642100133e-06,
|
| 414 |
+
"loss": 0.0378,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.6857142857142857,
|
| 419 |
+
"grad_norm": 0.019147174664357086,
|
| 420 |
+
"learning_rate": 3.66772795919611e-06,
|
| 421 |
+
"loss": 0.0329,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.7142857142857144,
|
| 426 |
+
"grad_norm": 0.023587816390011593,
|
| 427 |
+
"learning_rate": 3.044460665744284e-06,
|
| 428 |
+
"loss": 0.035,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.7428571428571429,
|
| 433 |
+
"grad_norm": 0.020512713984449708,
|
| 434 |
+
"learning_rate": 2.475778302439524e-06,
|
| 435 |
+
"loss": 0.0336,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.7714285714285714,
|
| 440 |
+
"grad_norm": 0.026037372894968262,
|
| 441 |
+
"learning_rate": 1.9630947032398067e-06,
|
| 442 |
+
"loss": 0.0326,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.8,
|
| 447 |
+
"grad_norm": 0.02608380575210609,
|
| 448 |
+
"learning_rate": 1.5076844803522922e-06,
|
| 449 |
+
"loss": 0.0415,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.8285714285714287,
|
| 454 |
+
"grad_norm": 0.0186002350742929,
|
| 455 |
+
"learning_rate": 1.1106798553464804e-06,
|
| 456 |
+
"loss": 0.0353,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.8571428571428572,
|
| 461 |
+
"grad_norm": 0.022985332334595585,
|
| 462 |
+
"learning_rate": 7.730678442730538e-07,
|
| 463 |
+
"loss": 0.0393,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.8857142857142857,
|
| 468 |
+
"grad_norm": 0.019579707985211765,
|
| 469 |
+
"learning_rate": 4.956878037864043e-07,
|
| 470 |
+
"loss": 0.0322,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.9142857142857141,
|
| 475 |
+
"grad_norm": 0.01775819492990442,
|
| 476 |
+
"learning_rate": 2.7922934437178695e-07,
|
| 477 |
+
"loss": 0.0355,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.9428571428571428,
|
| 482 |
+
"grad_norm": 0.01882371685418005,
|
| 483 |
+
"learning_rate": 1.2423061586496477e-07,
|
| 484 |
+
"loss": 0.0391,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.9714285714285715,
|
| 489 |
+
"grad_norm": 0.01746735435052931,
|
| 490 |
+
"learning_rate": 3.107696952694139e-08,
|
| 491 |
+
"loss": 0.0353,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.0,
|
| 496 |
+
"grad_norm": 0.018216407189787398,
|
| 497 |
+
"learning_rate": 0.0,
|
| 498 |
+
"loss": 0.0316,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 2.0,
|
| 503 |
+
"step": 70,
|
| 504 |
+
"total_flos": 382633893167104.0,
|
| 505 |
+
"train_loss": 0.07849941711340631,
|
| 506 |
+
"train_runtime": 603.8266,
|
| 507 |
+
"train_samples_per_second": 0.914,
|
| 508 |
+
"train_steps_per_second": 0.116
|
| 509 |
+
}
|
| 510 |
+
],
|
| 511 |
+
"logging_steps": 1,
|
| 512 |
+
"max_steps": 70,
|
| 513 |
+
"num_input_tokens_seen": 0,
|
| 514 |
+
"num_train_epochs": 2,
|
| 515 |
+
"save_steps": 500,
|
| 516 |
+
"stateful_callbacks": {
|
| 517 |
+
"TrainerControl": {
|
| 518 |
+
"args": {
|
| 519 |
+
"should_epoch_stop": false,
|
| 520 |
+
"should_evaluate": false,
|
| 521 |
+
"should_log": false,
|
| 522 |
+
"should_save": true,
|
| 523 |
+
"should_training_stop": true
|
| 524 |
+
},
|
| 525 |
+
"attributes": {}
|
| 526 |
+
}
|
| 527 |
+
},
|
| 528 |
+
"total_flos": 382633893167104.0,
|
| 529 |
+
"train_batch_size": 1,
|
| 530 |
+
"trial_name": null,
|
| 531 |
+
"trial_params": null
|
| 532 |
+
}
|